Merge branch 'collectd-4.8' into collectd-4.9
[collectd.git] / src / processes.c
index 7430afe..f2eb0a3 100644 (file)
@@ -4,6 +4,8 @@
  * Copyright (C) 2006-2008  Florian octo Forster
  * Copyright (C) 2008       Oleg King
  * Copyright (C) 2009       Sebastian Harl
+ * Copyright (C) 2009       Andrés J. Díaz
+ * Copyright (C) 2009       Manuel Sanmartin
  *
  * This program is free software; you can redistribute it and/or modify it
  * under the terms of the GNU General Public License as published by the
@@ -24,6 +26,8 @@
  *   Florian octo Forster <octo at verplant.org>
  *   Oleg King <king2 at kaluga.ru>
  *   Sebastian Harl <sh at tokkee.org>
+ *   Andrés J. Díaz <ajdiaz at connectical.com>
+ *   Manuel Sanmartin
  **/
 
 #include "collectd.h"
 #  include <sys/proc.h>
 /* #endif HAVE_LIBKVM_GETPROCS && HAVE_STRUCT_KINFO_PROC_FREEBSD */
 
+#elif HAVE_PROCINFO_H
+#  include <procinfo.h>
+#  include <sys/types.h>
+
+#define MAXPROCENTRY 32
+#define MAXTHRDENTRY 16
+#define MAXARGLN 1024
+/* #endif HAVE_PROCINFO_H */
+
 #else
 # error "No applicable input method."
 #endif
@@ -136,6 +149,12 @@ typedef struct procstat_entry_s
        unsigned long cpu_user_counter;
        unsigned long cpu_system_counter;
 
+       /* io data */
+       derive_t io_rchar;
+       derive_t io_wchar;
+       derive_t io_syscr;
+       derive_t io_syscw;
+
        struct procstat_entry_s *next;
 } procstat_entry_t;
 
@@ -159,6 +178,12 @@ typedef struct procstat
        unsigned long cpu_user_counter;
        unsigned long cpu_system_counter;
 
+       /* io data */
+       derive_t io_rchar;
+       derive_t io_wchar;
+       derive_t io_syscr;
+       derive_t io_syscw;
+
        struct procstat   *next;
        struct procstat_entry_s *instances;
 } procstat_t;
@@ -179,7 +204,19 @@ static long pagesize_g;
 
 #elif HAVE_LIBKVM_GETPROCS && HAVE_STRUCT_KINFO_PROC_FREEBSD
 /* no global variables */
-#endif /* HAVE_LIBKVM_GETPROCS && HAVE_STRUCT_KINFO_PROC_FREEBSD */
+/* #endif HAVE_LIBKVM_GETPROCS && HAVE_STRUCT_KINFO_PROC_FREEBSD */
+
+#elif HAVE_PROCINFO_H
+static  struct procentry64 procentry[MAXPROCENTRY];
+static  struct thrdentry64 thrdentry[MAXTHRDENTRY];
+static int pagesize;
+
+#ifndef _AIXVERSION_610
+int     getprocs64 (void *procsinfo, int sizproc, void *fdsinfo, int sizfd, pid_t *index, int count);
+int     getthrds64( pid_t, void *, int, tid64_t *, int );
+#endif
+int getargs (struct procentry64 *processBuffer, int bufferLen, char *argsBuffer, int argsLen);
+#endif /* HAVE_PROCINFO_H */
 
 /* put name of process from config to list_head_g tree
    list_head_g is a list of 'procstat_t' structs with
@@ -231,7 +268,7 @@ static void ps_list_register (const char *name, const char *regexp)
                return;
        }
 #endif
-       
+
        for (ptr = list_head_g; ptr != NULL; ptr = ptr->next)
        {
                if (strcmp (ptr->name, name) == 0)
@@ -307,13 +344,13 @@ static void ps_list_add (const char *name, const char *cmdline, procstat_entry_t
                if ((pse == NULL) || (pse->id != entry->id))
                {
                        procstat_entry_t *new;
-                       
+
                        new = (procstat_entry_t *) malloc (sizeof (procstat_entry_t));
                        if (new == NULL)
                                return;
                        memset (new, 0, sizeof (procstat_entry_t));
                        new->id = entry->id;
-                       
+
                        if (pse == NULL)
                                ps->instances = new;
                        else
@@ -328,6 +365,10 @@ static void ps_list_add (const char *name, const char *cmdline, procstat_entry_t
                pse->vmem_size  = entry->vmem_size;
                pse->vmem_rss   = entry->vmem_rss;
                pse->stack_size = entry->stack_size;
+               pse->io_rchar   = entry->io_rchar;
+               pse->io_wchar   = entry->io_wchar;
+               pse->io_syscr   = entry->io_syscr;
+               pse->io_syscw   = entry->io_syscw;
 
                ps->num_proc   += pse->num_proc;
                ps->num_lwp    += pse->num_lwp;
@@ -335,6 +376,11 @@ static void ps_list_add (const char *name, const char *cmdline, procstat_entry_t
                ps->vmem_rss   += pse->vmem_rss;
                ps->stack_size += pse->stack_size;
 
+               ps->io_rchar   += ((pse->io_rchar == -1)?0:pse->io_rchar);
+               ps->io_wchar   += ((pse->io_wchar == -1)?0:pse->io_wchar);
+               ps->io_syscr   += ((pse->io_syscr == -1)?0:pse->io_syscr);
+               ps->io_syscw   += ((pse->io_syscw == -1)?0:pse->io_syscw);
+
                if ((entry->vmem_minflt_counter == 0)
                                && (entry->vmem_majflt_counter == 0))
                {
@@ -356,7 +402,7 @@ static void ps_list_add (const char *name, const char *cmdline, procstat_entry_t
                                pse->vmem_minflt = entry->vmem_minflt_counter - pse->vmem_minflt_counter;
                        }
                        pse->vmem_minflt_counter = entry->vmem_minflt_counter;
-                       
+
                        if (entry->vmem_majflt_counter < pse->vmem_majflt_counter)
                        {
                                pse->vmem_majflt = entry->vmem_majflt_counter
@@ -393,7 +439,7 @@ static void ps_list_add (const char *name, const char *cmdline, procstat_entry_t
                                pse->cpu_user = entry->cpu_user_counter - pse->cpu_user_counter;
                        }
                        pse->cpu_user_counter = entry->cpu_user_counter;
-                       
+
                        if (entry->cpu_system_counter < pse->cpu_system_counter)
                        {
                                pse->cpu_system = entry->cpu_system_counter
@@ -425,6 +471,10 @@ static void ps_list_reset (void)
                ps->vmem_size   = 0;
                ps->vmem_rss    = 0;
                ps->stack_size  = 0;
+               ps->io_rchar = -1;
+               ps->io_wchar = -1;
+               ps->io_syscr = -1;
+               ps->io_syscw = -1;
 
                pse_prev = NULL;
                pse = ps->instances;
@@ -538,7 +588,11 @@ static int ps_init (void)
 
 #elif HAVE_LIBKVM_GETPROCS && HAVE_STRUCT_KINFO_PROC_FREEBSD
 /* no initialization */
-#endif /* HAVE_LIBKVM_GETPROCS && HAVE_STRUCT_KINFO_PROC_FREEBSD */
+/* #endif HAVE_LIBKVM_GETPROCS && HAVE_STRUCT_KINFO_PROC_FREEBSD */
+
+#elif HAVE_PROCINFO_H
+       pagesize = getpagesize();
+#endif /* HAVE_PROCINFO_H */
 
        return (0);
 } /* int ps_init */
@@ -607,12 +661,33 @@ static void ps_submit_proc_list (procstat_t *ps)
        vl.values_len = 2;
        plugin_dispatch_values (&vl);
 
+       if ( (ps->io_rchar != -1) && (ps->io_wchar != -1) )
+       {
+               sstrncpy (vl.type, "ps_disk_octets", sizeof (vl.type));
+               vl.values[0].derive = ps->io_rchar;
+               vl.values[1].derive = ps->io_wchar;
+               vl.values_len = 2;
+               plugin_dispatch_values (&vl);
+       }
+
+       if ( (ps->io_syscr != -1) && (ps->io_syscw != -1) )
+       {
+               sstrncpy (vl.type, "ps_disk_ops", sizeof (vl.type));
+               vl.values[0].derive = ps->io_syscr;
+               vl.values[1].derive = ps->io_syscw;
+               vl.values_len = 2;
+               plugin_dispatch_values (&vl);
+       }
+
        DEBUG ("name = %s; num_proc = %lu; num_lwp = %lu; vmem_rss = %lu; "
                        "vmem_minflt_counter = %lu; vmem_majflt_counter = %lu; "
-                       "cpu_user_counter = %lu; cpu_system_counter = %lu;",
+                       "cpu_user_counter = %lu; cpu_system_counter = %lu; "
+                       "io_rchar = %"PRIi64"; io_wchar = %"PRIi64"; "
+                       "io_syscr = %"PRIi64"; io_syscw = %"PRIi64";",
                        ps->name, ps->num_proc, ps->num_lwp, ps->vmem_rss,
                        ps->vmem_minflt_counter, ps->vmem_majflt_counter,
-                       ps->cpu_user_counter, ps->cpu_system_counter);
+                       ps->cpu_user_counter, ps->cpu_system_counter,
+                       ps->io_rchar, ps->io_wchar, ps->io_syscr, ps->io_syscw);
 } /* void ps_submit_proc_list */
 
 /* ------- additional functions for KERNEL_LINUX/HAVE_THREAD_INFO ------- */
@@ -644,6 +719,60 @@ static int ps_read_tasks (int pid)
        return ((count >= 1) ? count : 1);
 } /* int *ps_read_tasks */
 
+static procstat_t *ps_read_io (int pid, procstat_t *ps)
+{
+       FILE *fh;
+       char buffer[1024];
+       char filename[64];
+
+       char *fields[8];
+       int numfields;
+
+       ssnprintf (filename, sizeof (filename), "/proc/%i/io", pid);
+       if ((fh = fopen (filename, "r")) == NULL)
+               return (NULL);
+
+       while (fgets (buffer, 1024, fh) != NULL)
+       {
+               derive_t *val = NULL;
+               long long tmp;
+               char *endptr;
+
+               if (strncasecmp (buffer, "rchar:", 6) == 0)
+                       val = &(ps->io_rchar);
+               else if (strncasecmp (buffer, "wchar:", 6) == 0)
+                       val = &(ps->io_wchar);
+               else if (strncasecmp (buffer, "syscr:", 6) == 0)
+                       val = &(ps->io_syscr);
+               else if (strncasecmp (buffer, "syscw:", 6) == 0)
+                       val = &(ps->io_syscw);
+               else
+                       continue;
+
+               numfields = strsplit (buffer, fields, 8);
+
+               if (numfields < 2)
+                       continue;
+
+               errno = 0;
+               endptr = NULL;
+               tmp = strtoll (fields[1], &endptr, /* base = */ 10);
+               if ((errno != 0) || (endptr == fields[1]))
+                       *val = -1;
+               else
+                       *val = (derive_t) tmp;
+       } /* while (fgets) */
+
+       if (fclose (fh))
+       {
+               char errbuf[1024];
+               WARNING ("processes: fclose: %s",
+                               sstrerror (errno, errbuf, sizeof (errbuf)));
+       }
+
+       return (ps);
+} /* procstat_t *ps_read_io */
+
 int ps_read_process (int pid, procstat_t *ps, char *state)
 {
        char  filename[64];
@@ -746,6 +875,17 @@ int ps_read_process (int pid, procstat_t *ps, char *state)
        ps->vmem_rss = (unsigned long) vmem_rss;
        ps->stack_size = (unsigned long) stack_size;
 
+       if ( (ps_read_io (pid, ps)) == NULL)
+       {
+               /* no io data */
+               ps->io_rchar = -1;
+               ps->io_wchar = -1;
+               ps->io_syscr = -1;
+               ps->io_syscw = -1;
+
+               DEBUG("ps_read_process: not get io data for pid %i",pid);
+       }
+
        /* success */
        return (0);
 } /* int ps_read_process (...) */
@@ -839,6 +979,70 @@ static char *ps_get_cmdline (pid_t pid, char *name, char *buf, size_t buf_len)
        }
        return buf;
 } /* char *ps_get_cmdline (...) */
+
+static unsigned long read_fork_rate ()
+{
+       FILE *proc_stat;
+       char buf[1024];
+       unsigned long result = 0;
+       int numfields;
+       char *fields[3];
+
+       proc_stat = fopen("/proc/stat", "r");
+       if (proc_stat == NULL) {
+               char errbuf[1024];
+               ERROR ("processes plugin: fopen (/proc/stat) failed: %s",
+                               sstrerror (errno, errbuf, sizeof (errbuf)));
+               return ULONG_MAX;
+       }
+
+       while (fgets (buf, sizeof(buf), proc_stat) != NULL)
+       {
+               char *endptr;
+
+               numfields = strsplit(buf, fields, STATIC_ARRAY_SIZE (fields));
+               if (numfields != 2)
+                       continue;
+
+               if (strcmp ("processes", fields[0]) != 0)
+                       continue;
+
+               errno = 0;
+               endptr = NULL;
+               result = strtoul(fields[1], &endptr, 10);
+               if ((endptr == fields[1]) || (errno != 0)) {
+                       ERROR ("processes plugin: Cannot parse fork rate: %s",
+                                       fields[1]);
+                       result = ULONG_MAX;
+                       break;
+               }
+
+               break;
+       }
+
+       fclose(proc_stat);
+
+       return result;
+}
+
+static void ps_submit_fork_rate (unsigned long value)
+{
+       value_t values[1];
+       value_list_t vl = VALUE_LIST_INIT;
+
+       values[0].derive = (derive_t) value;
+
+       vl.values = values;
+       vl.values_len = 1;
+       sstrncpy (vl.host, hostname_g, sizeof (vl.host));
+       sstrncpy (vl.plugin, "processes", sizeof (vl.plugin));
+       sstrncpy (vl.plugin_instance, "", sizeof (vl.plugin_instance));
+       sstrncpy (vl.type, "fork_rate", sizeof (vl.type));
+       sstrncpy (vl.type_instance, "", sizeof (vl.type_instance));
+
+       plugin_dispatch_values (&vl);
+}
+
 #endif /* KERNEL_LINUX */
 
 #if HAVE_THREAD_INFO
@@ -1161,6 +1365,8 @@ static int ps_read (void)
        procstat_entry_t pse;
        char       state;
 
+       unsigned long fork_rate;
+
        procstat_t *ps_ptr;
 
        running = sleeping = zombies = stopped = paging = blocked = 0;
@@ -1208,6 +1414,11 @@ static int ps_read (void)
                pse.cpu_system = 0;
                pse.cpu_system_counter = ps.cpu_system_counter;
 
+               pse.io_rchar = ps.io_rchar;
+               pse.io_wchar = ps.io_wchar;
+               pse.io_syscr = ps.io_syscr;
+               pse.io_syscw = ps.io_syscw;
+
                switch (state)
                {
                        case 'R': running++;  break;
@@ -1234,6 +1445,10 @@ static int ps_read (void)
 
        for (ps_ptr = list_head_g; ps_ptr != NULL; ps_ptr = ps_ptr->next)
                ps_submit_proc_list (ps_ptr);
+
+       fork_rate = read_fork_rate();
+       if (fork_rate != ULONG_MAX)
+               ps_submit_fork_rate(fork_rate);
 /* #endif KERNEL_LINUX */
 
 #elif HAVE_LIBKVM_GETPROCS && HAVE_STRUCT_KINFO_PROC_FREEBSD
@@ -1332,6 +1547,12 @@ static int ps_read (void)
                        * 1000
                        + procs[i].ki_rusage.ru_stime.tv_usec;
 
+               /* no io data */
+               pse.io_rchar = -1;
+               pse.io_wchar = -1;
+               pse.io_syscr = -1;
+               pse.io_syscw = -1;
+
                switch (procs[i].ki_stat)
                {
                        case SSTOP:     stopped++;      break;
@@ -1358,7 +1579,133 @@ static int ps_read (void)
 
        for (ps_ptr = list_head_g; ps_ptr != NULL; ps_ptr = ps_ptr->next)
                ps_submit_proc_list (ps_ptr);
-#endif /* HAVE_LIBKVM_GETPROCS && HAVE_STRUCT_KINFO_PROC_FREEBSD */
+/* #endif HAVE_LIBKVM_GETPROCS && HAVE_STRUCT_KINFO_PROC_FREEBSD */
+
+#elif HAVE_PROCINFO_H
+       /* AIX */
+       int running  = 0;
+       int sleeping = 0;
+       int zombies  = 0;
+       int stopped  = 0;
+       int paging   = 0;
+       int blocked  = 0;
+
+       pid_t pindex = 0;
+       int nprocs;
+
+       procstat_t *ps;
+       procstat_entry_t pse;
+
+       ps_list_reset ();
+       while ((nprocs = getprocs64 (procentry, sizeof(struct procentry64),
+                                       /* fdsinfo = */ NULL, sizeof(struct fdsinfo64),
+                                       &pindex, MAXPROCENTRY)) > 0)
+       {
+               int i;
+
+               for (i = 0; i < nprocs; i++)
+               {
+                       tid64_t thindex;
+                       int nthreads;
+                       char arglist[MAXARGLN+1];
+                       char *cargs;
+                       char *cmdline;
+
+                       if (procentry[i].pi_state == SNONE) continue;
+                       /* if (procentry[i].pi_state == SZOMB)  FIXME */
+
+                       cmdline = procentry[i].pi_comm;
+                       cargs = procentry[i].pi_comm;
+                       if ( procentry[i].pi_flags & SKPROC )
+                       {
+                               if (procentry[i].pi_pid == 0)
+                                       cmdline = "swapper";
+                               cargs = cmdline;
+                       }
+                       else
+                       {
+                               if (getargs(&procentry[i], sizeof(struct procentry64), arglist, MAXARGLN) >= 0)
+                               {
+                                       int n;
+
+                                       n = -1;
+                                       while (++n < MAXARGLN)
+                                       {
+                                               if (arglist[n] == '\0')
+                                               {
+                                                       if (arglist[n+1] == '\0')
+                                                               break;
+                                                       arglist[n] = ' ';
+                                               }
+                                       }
+                                       cargs = arglist;
+                               }
+                       }
+
+                       pse.id       = procentry[i].pi_pid;
+                       pse.age      = 0;
+                       pse.num_lwp  = procentry[i].pi_thcount;
+                       pse.num_proc = 1;
+
+                       thindex=0;
+                       while ((nthreads = getthrds64(procentry[i].pi_pid,
+                                                       thrdentry, sizeof(struct thrdentry64),
+                                                       &thindex, MAXTHRDENTRY)) > 0)
+                       {
+                               int j;
+
+                               for (j=0; j< nthreads; j++)
+                               {
+                                       switch (thrdentry[j].ti_state)
+                                       {
+                                               /* case TSNONE: break; */
+                                               case TSIDL:     blocked++;      break; /* FIXME is really blocked */
+                                               case TSRUN:     running++;      break;
+                                               case TSSLEEP:   sleeping++;     break;
+                                               case TSSWAP:    paging++;       break;
+                                               case TSSTOP:    stopped++;      break;
+                                               case TSZOMB:    zombies++;      break;
+                                       }
+                               }
+                               if (nthreads < MAXTHRDENTRY)
+                                       break;
+                       }
+
+                       pse.cpu_user = 0;
+                       /* tv_usec is nanosec ??? */
+                       pse.cpu_user_counter = procentry[i].pi_ru.ru_utime.tv_sec * 1000000 +
+                               procentry[i].pi_ru.ru_utime.tv_usec / 1000;
+
+                       pse.cpu_system = 0;
+                       /* tv_usec is nanosec ??? */
+                       pse.cpu_system_counter = procentry[i].pi_ru.ru_stime.tv_sec * 1000000 +
+                               procentry[i].pi_ru.ru_stime.tv_usec / 1000;
+
+                       pse.vmem_minflt = 0;
+                       pse.vmem_minflt_counter = procentry[i].pi_minflt;
+                       pse.vmem_majflt = 0;
+                       pse.vmem_majflt_counter = procentry[i].pi_majflt;
+
+                       pse.vmem_size = procentry[i].pi_tsize + procentry[i].pi_dvm * pagesize;
+                       pse.vmem_rss = (procentry[i].pi_drss + procentry[i].pi_trss) * pagesize;
+                       pse.stack_size =  0;
+
+                       ps_list_add (cmdline, cargs, &pse);
+               } /* for (i = 0 .. nprocs) */
+
+               if (nprocs < MAXPROCENTRY)
+                       break;
+       } /* while (getprocs64() > 0) */
+       ps_submit_state ("running",  running);
+       ps_submit_state ("sleeping", sleeping);
+       ps_submit_state ("zombies",  zombies);
+       ps_submit_state ("stopped",  stopped);
+       ps_submit_state ("paging",   paging);
+       ps_submit_state ("blocked",  blocked);
+
+       for (ps = list_head_g; ps != NULL; ps = ps->next)
+               ps_submit_proc_list (ps);
+#endif /* HAVE_PROCINFO_H */
 
        return (0);
 } /* int ps_read */